EP0672999A1 - Pixelkorrectur und Glättenverfahren - Google Patents

Pixelkorrectur und Glättenverfahren Download PDF

Info

Publication number
EP0672999A1
EP0672999A1 EP95301537A EP95301537A EP0672999A1 EP 0672999 A1 EP0672999 A1 EP 0672999A1 EP 95301537 A EP95301537 A EP 95301537A EP 95301537 A EP95301537 A EP 95301537A EP 0672999 A1 EP0672999 A1 EP 0672999A1
Authority
EP
European Patent Office
Prior art keywords
pixels
pixel
type
high resolution
bit map
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
EP95301537A
Other languages
English (en)
French (fr)
Other versions
EP0672999B1 (de
Inventor
Mark A. Overton
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
HP Inc
Original Assignee
Hewlett Packard Co
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hewlett Packard Co filed Critical Hewlett Packard Co
Publication of EP0672999A1 publication Critical patent/EP0672999A1/de
Application granted granted Critical
Publication of EP0672999B1 publication Critical patent/EP0672999B1/de
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/20—Drawing from basic elements, e.g. lines or circles
    • G06T11/203—Drawing of straight lines or curves

Definitions

  • This invention relates to the printing of pixels on a recording medium and, in particular, to a method for improving the quality of a displayed (e.g., printed) pattern formed of pixels.
  • Text and pictures are sometimes printed using closely spaced dots, called pixels.
  • the individual pixels are not easily perceived by the human eye but rather appear to merge into adjacent pixels to form a continuous, solid symbol or image.
  • Such pixels are printed by facsimile machines, dot matrix printers, and other printing or displaying devices.
  • pixels being digital rather than analog, inherently results in some quantization error, since a pixel is usually either black or white. Pixels of complementary colors or gray shades can also be printed, and the resulting printed patterns suffer (to varying degrees) from the same quantization effects as patterns printed using only black and white pixels. Thus, a diagonal line printed with pixels will usually appear as small steps rather than a smooth diagonal line due to this quantization error. This problem also appears in attempting to print any symbol or line which is not comprised of solely horizontal or vertical lines. As the printed dots per inch (dpi) is increased, this quantization error becomes less noticeable.
  • dpi dots per inch
  • the dpi printing in facsimile machines is intentionally set to be relatively low to enable faster transmission of the pixel data to the receiving facsimile machine.
  • diagonal and rounded figures appear jagged to the naked eye.
  • One known amelioration technique is to detect a certain pixel pattern in a block of pixels and insert an interpolated row of pixels to provide smoothing of this pixel pattern.
  • Such smoothing may be performed by providing a look-up table which is addressed by the incoming pixel pattern and whose output contains one or more interpolated rows of pixels to augment the original pattern.
  • the patterns stored in the look-up table can themselves be flawed and are difficult to generate.
  • Another method for smoothing a pixel pattern is to perform low pass filtering on the pixel pattern. This technique removes stray dots and ticks as well as smooths the steps in diagonal or curved lines; however, such low pass filtering may also eliminate details from the original image, such as narrow white or black lines. Thus, such image processing causes dark and light areas of pictures to be unevenly affected, resulting in distortion.
  • the above techniques are usually used in combination with pixel scaling, where the dpi of the printer is set to be greater than the dpi of the original pixel pattern to increase the resolution of the printed pattern.
  • a block of four pixels may be printed for every one original pixel.
  • Each printed pixel would then be approximately one-quarter the size of the original pixel to effectively increase the resolution of the printed pattern by four.
  • Such scaling only has advantages when some correction of the original pixel pattern has been performed.
  • This method first places the original pixel pattern in a bit mapped memory.
  • a small block of this bit mapped pattern acts as an address for a look-up table containing a corrected pixel pattern portion (an output tile) corresponding to the target pixels to be corrected in the input tile.
  • the input tile consists of two adjacent target pixels and their immediate neighboring pixels to form an input tile of twelve pixels.
  • the corresponding output tile is then printed. at a higher dots-per-inch (dpi) than the original pixel pattern to provide a corrected and higher resolution pattern.
  • This process may also be performed using an algorithm which performs a correction on the target pixels in the input tile instead of addressing a corresponding output tile in a look-up table. The above process is repeated until all of the original pixels have had output tiles generated for them.
  • a novel method is described for deriving the output tiles.
  • a high resolution bit map is created from an input tile by scaling the input tile by, for example, sixteen times in the x and y directions.
  • a triangular portion of each convex corner in the high resolution bit map is chopped off, and a triangular portion of each concave corner in the high resolution bit map is filled in.
  • the size of these triangular portions is determined using a function based on the exposed faces per minority pixel in the input tile.
  • An exposed face of a pixel is a white-black (or black/white) transition on one of the pixel's sides.
  • a minority pixel is whichever pixel type (e.g., black pixel or white pixel) is fewest in number in the input tile.
  • the quotient of exposed faces divided by the number of minority pixels in the input tile is a measure of the average vulnerability of pixels in the input tile to being eliminated by the fills and chops in the above-mentioned step. The higher the quotient, the less smoothing should be done in this fills and chops step. This regulation of smoothing helps preserve desired detail in the image. Other suitable quotients representative of the vulnerability of pixels to being eliminated in a subsequent smoothing process may also be used to regulate smoothing.
  • Low pass filtering is then performed on the resulting high resolution bit map.
  • the quotient previously calculated also regulates the amount of low pass filtering performed. This low pass filtering is much less than that needed by the prior art low pass filtering methods, since the high resolution bit map has already been partially smoothed in the previous step.
  • the middle portion of the high resolution bit map corresponding to the two original target pixels intended to be corrected is then divided up into a number of pixel areas, where each area represents a pixel to be printed.
  • the middle portion of the bit map is divided into six pixel areas. Due to the initial scaling by sixteen, a large number of pixels resides in each pixel area.
  • a threshold number is set which determines whether a pixel area should be printed as a black dot or a white dot. This threshold may be adjusted based upon the amount of ink-bleed, or other marking agent bleed, experienced when the image is printed on a recording medium.
  • the preferred method also detects input tile exceptions which are not subject to the normal corrections.
  • Such an exception may be the detection of a stray dot or a notch or tick in a line of pixels. These exceptions are then corrected accordingly.
  • a method for detecting whether a pixel pattern is text or an image is also described.
  • Fig. 1 is a block diagram of a circuit which may be used to perform the inventive method.
  • Fig. 2 is a flow chart illustrating the preferred steps in a method for creating output tiles.
  • Figs. 3A, 3B, 3C, and 3D illustrate the processing of an input tile to generate an output tile having a corrected pixel pattern.
  • Fig. 4 is a flow chart illustrating the steps used to fetch an output tile from a look-up table.
  • Fig. 5 illustrates a correction made to a stray dot in an input tile by eliminating the stray dot.
  • Fig. 6 illustrates an exception when a pixel pattern includes a tick, where the tick has been removed but is reflected in the outputted pixel pattern.
  • Fig. 7 illustrates an exception when a pixel pattern includes a notch, where the notch has been removed but is reflected in the outputted pixel pattern.
  • Fig. 8 is a flow chart illustrating the steps used to determine whether an input tile is a part of an image or text.
  • Fig. 1 is a block diagram illustrating one embodiment of the invention. The operation of the inventive structure will be described in the context of processing pixel data received at a facsimile machine containing the circuitry of Fig. 1. This invention, however, may also be applied to any type of display device using pixel type displaying where the original pixel pattern is desired to be enhanced and outputted at a higher resolution than the original pixel pattern.
  • the raw facsimile data is received at an input port of a conventional modem and suitably decompressed to convert the transmitted digital codes into pixels for printing by the receiving facsimile machine.
  • This conventional circuitry is shown as modem/decompressor 10.
  • This processed pixel data is then temporarily stored in a first bit mapped memory 12.
  • the pixels in bit mapped memory 12 would be printed by the facsimile machine.
  • This pixel image contains perceptible quantization errors and other undesirable qualities. To remove these undesirable qualities and to enhance the pixel data stored in bit mapped memory 12, the following circuitry is provided.
  • a processor 14 is used to modify the pixel data contained in memory 12 in accordance with a set of rules.
  • the processor 14 operates by receiving an input tile from bit mapped memory 12 and outputting a corresponding output tile to bit mapped memory 20.
  • processor 14 performs an algorithm on an input tile of the pixel data, in accordance with a set of instructions contained in a program ROM 16, to convert target pixels in the input tile to an output tile containing enhanced pixel data.
  • processor 14 may be an ASIC which is hard-wired to replace target pixels in the input tile with an output tile previously stored in a look-up table 18.
  • a microprocessor may also be used to index output tiles in look-up table 18. The use of an algorithm and the look-up table 18 are described in detail below.
  • step 1 the processor 14 addresses a small block (an input tile) of pixel data in memory 12.
  • one input tile consists of two adjacent pixels in a row in memory 12 and their nearest neighbors.
  • One such input tile 28 is shown in Fig. 3A as comprising twelve pixels or bits, where the central two pixels 29, 30 are the target pixels which are to be enhanced by the processor 14.
  • the pixel data stored in memory 12 has an original resolution of N dots per inch (dpi) in the horizontal direction and M dpi in the vertical direction.
  • step 2 processor 14 determines whether this input tile 28 is an exception.
  • An exception occurs when the input tile matches one of a number of specific patterns which require special processing other than that described in the remainder of the flowchart of Fig. 2.
  • One such exception is the appearance of a single isolated bit in the input tile, as illustrated in Fig. 5. This bit is assumed to be a stray bit appearing as a result of noise or a "dirty" original document. This bit is not enhanced but is simply deleted in a corresponding output tile (step 2'). Examples of other exceptions and the method used to handle these exceptions will be described later.
  • a high resolution bit map 31 (Fig. 3B) is created from the input tile 28.
  • this high resolution bit map 31 is formed by expanding the input tile 28 by a factor of sixteen in both the x and y directions, as shown in Fig. 3B.
  • the array of black pixels created by expanding input tile 28 is enclosed within the solid outline 32.
  • a factor used to expand the input tile 28 in the x direction is different than a factor used to expand the tile 28 in the y direction.
  • step 4 the processor 14 determines whether the high resolution bit map 31 is part of an image (e.g., a picture) or text. If it is determined that the bit map 31 is part of an image, the smoothing steps 5-7 are bypassed and step 8 is then performed on the bit map 31. The method for determining whether the bit map 31 is part of an image or text will be described later. If it is determined that the bit map 31 is part of text, the following operations are performed to improve the quality of the printed text.
  • step 5 the following calculation is performed to determine the extent of initial smoothing to be conducted in step 6:
  • Q exposed minority pixel faces in input tile number of minority pixels in input tile where a higher Q value results in less smoothing being performed in steps 6 and 7.
  • the exposed pixel faces referred to in Equation 1 is the number of faces (maximum of four per pixel) of a minority pixel facing a white-black or black-white transition; the borders of the input tile are not counted as a transition.
  • a minority pixel is the pixel type (black or white) which is fewest in number in the input tile.
  • the Q value is thus the average number of exposed faces per minority pixel in the input tile.
  • This Q value is a measure of the average vulnerability of pixels to being eliminated by the initial smoothing in step 6.
  • a correction factor i.e., the Q value
  • Regulating the amount of smoothing not only helps preserve detail, but also lessens the damage done if the text/image separation step 4, mentioned above, fails. If an input tile is erroneously considered text (but is actually part of an image), the pixels in it are likely to be somewhat scattered, resulting in a high Q value and causing less smoothing to occur. The picture is then not significantly altered.
  • This Q value is also used in determining the size of the template used in performing low pass digital filtering (step 7) on the high resolution bit map after step 6 has been concluded.
  • Suitable Q values may be calculated by using equations other than Equation 1.
  • another suitable Q value may be calculated as follows: compute the number of exposed corners on the most exposed middle pixel, corresponding to target pixel 29 or 30 (Fig. 3A), in an input tile.
  • the "exposure" of a Pixel is the number of exposed corners it has.
  • a pixel's corner is "exposed” if its three neighbors are all the opposite color or type. For example, assume in the arrangement of pixels below that the B pixels are black and the W pixels are white.
  • the B0 pixel above has one exposed corner (i.e., its upper right corner.
  • the B1 pixel has two exposed corners.
  • the exposed corners of the most exposed pixel is then stored as the Q value. This "exposure” is a number from zero to four.
  • This exposure number can be used instead of the Q value specifically identified in Equation 1.
  • the exposure of the most exposed pixel as computed above is one, and if either both the top and bottom sides of the pixel are exposed or both the left and right sides are exposed, then the exposure is increased to two.
  • the actual exposure of pixel B0 above is one, but because both the top and bottom of B0 are exposed, the exposure is increased to two, reflecting the fact that this pixel B0 is more vulnerable to being eliminated than its one exposed corner would suggest.
  • step 6 an initial smoothing of the 900 corners of white/black or black/white pixel transitions in the high resolution bit map 31 is performed by either chopping off a certain size triangular corner 34 (Fig. 3C) of the 90° convex angles or filling in the 90° concave angles with a certain size triangle 35.
  • the size of these triangles corresponds to the previously determined Q value, where a higher value of Q corresponds to smaller triangles (i.e., less smoothing) being used in the chopping and filling step above.
  • Such a chopping and filling step is illustrated in Fig. 3C.
  • the pixels deleted or added in step 6 are shown within dashed lines.
  • Table I illustrates one example of a correlation between a calculated Q value and the percentage of the maximum triangle size to be chopped off or filled during step 6.
  • Table I O value Chop/Fill % of Max 0 - .99 100 1 - 1.99 100 2 - 2.99 75 3 - 4 35
  • a large initial pre-smoothing using the triangular chops and fills is performed by, for example, chopping off a right triangle portion having twelve pixels on two of its equal sides from the convex corners and filling in the concave corner with a right triangle having twelve pixels on two of its equal sides.
  • the maximum triangle size to be used in step 6 can be determined empirically and depends upon the original dpi of the pixel pattern and the degree of initial smoothing desired.
  • step 7 the resulting pre-smoothed high resolution bit map 31 is subjected to a conventional digital convolution using a relatively small convolution template which acts as a low-pass filter.
  • This filter is designed to smooth small rough edges but still leave the larger features mostly unchanged.
  • the digital filter in step 7 can use a higher-pass convolution than the prior art filters so as to preserve any thin black lines and thin white gaps.
  • smoothing is performed, yet details are not obliterated.
  • the larger the Q value the smaller the template used for digital filtering.
  • the resulting high resolution bit map 31 after smoothing by the conventional digital filter is illustrated by the outline of the black pixels in Fig. 3D.
  • the middle portion of the high resolution bit map 31 corresponding to the two target pixels 29, 30 in the input tile 28 is segmented into as many pixel areas 38-43 (Fig. 3D) as are desired to replace the two original target pixels 29, 30.
  • the area of the high resolution bit map 31 corresponding to the two target pixels 29, 30 would be divided up into six equal size pixel areas 38-43. Pixel areas 38-43 may overlap to some degree if desired.
  • the horizontal scaling for printing is 1.5 (corresponding to three output pixels in a row) and the vertical scaling is two (corresponding to two rows of output pixels).
  • the vertical scaling is three so that the dpi in both the horizontal and vertical directions are the same (e.g., 300 dpi).
  • step 9 the number of black pixels within a single pixel area 38-43 is determined and, if this number is greater than a threshold number, the pixel to be printed corresponding to that area 38-43 will be a black pixel. If the number of black pixels is below the threshold, a white pixel will be printed for that pixel area.
  • This threshold number can also be set based on the number of white pixels in a pixel area.
  • the threshold number may be adjusted to take into account any ink bleed, or other marking agent bleed, when printing or otherwise displaying a pixel on a medium. By properly setting the threshold number, a faithful reproduction of the overall image brightness or darkness can be had.
  • the threshold number may be set based upon empirical methods in view of the type of marking agent used, the recording medium used, and the printing method used. In the example of Fig. 3D, the threshold number is set to one half the total number of pixels in a pixel area 38-43, and the resulting determination of a black or white pixel to be printed for each pixel area 38-43 is shown by a B or W.
  • the output tile 46 (Fig. 3D) consisting of one pixel per pixel area 38-43 is outputted from the processor 14 and stored in a second bit mapped memory 20 (Fig. 1) for subsequent display (e.g., printing) on a medium.
  • the generated output tiles may instead be stored in look-up table 18 for later indexing by a corresponding input tile.
  • the output tile contains six pixels for every two target pixels 29, 30, the bit mapped memory 20 is three times as large as the bit mapped memory 12.
  • the above process is performed on all or a number of input tiles in the bit mapped memory 12 to generate a number of output tiles 46 stored in bit mapped memory 20.
  • the pixels in the output tiles 46 stored in bit mapped memory 20 are now sequentially addressed and printed (or otherwise displayed) in step 11 at a horizontal dpi of U and a vertical dpi of V, which are higher than the original horizontal dpi and vertical dpi of N and M, respectively.
  • six pixels will be printed in the space of two pixels from the original pixel pattern stored in memory 12, although any other size output tile (e.g., a 3 x 3 array) may be generated, stored, and displayed.
  • Conventional addressing and printhead driver circuitry 22 (Fig. 1) is used to output the pixels from memory 20.
  • Printhead driver 22 controls a conventional printhead 24 to print or otherwise display the corresponding pixels on a recording or display medium.
  • any number of pixels may be contained in an output tile, depending upon the desired resolution of the final printed pixel pattern. Further, any level of smoothing may be performed in steps 6 and 7, and this optimum smoothing may be determined empirically, depending upon what is believed by the user to be the most pleasing effect.
  • look-up table 18 contains the output tiles for every combination of twelve pixels in an input tile, for a total of 4096 output tiles.
  • the output tiles are initially created for all possible input tiles using the procedure described above with respect to Fig. 2.
  • the input tiles then act as an index to select a corresponding output tile from look-up table 18, and this indexed output tile is then outputted to bit mapped memory 20.
  • step 1 corresponds to step 1 in Fig. 2.
  • step 2 an output tile in look-up table 18 is addressed or indexed using the input tile as the address.
  • the input tile 28 in Fig. 3A would address the stored output tile 46 in Fig. 3D.
  • this output tile is transferred to bit mapped memory 20.
  • the pixels in bit mapped memory 20 are printed out at a horizontal dpi of U and a vertical dpi of V, which are higher than the original horizontal and vertical dpi of N and M, respectively.
  • the output tiles stored in look-up table 18 are derived using the process illustrated in Figs. 2 and 3A-3D for each input tile desired to be enhanced.
  • the input tile exceptions include a stray dot exception and a tick or notch exception.
  • the stray dot exception is illustrated in Fig. 5, where the input tile 52 contains an isolated black or white pixel. Since there are no intentional characters being formed of a single pixel, it is presumed that this isolated pixel is a product of noise or a dirty original. Accordingly, once this type of input tile is detected, the isolated pixel is deleted, and a corrected output tile 54 is outputted by processor 14.
  • the particular illustration of Fig. 5 shows the center two target pixels being converted into a 3 x 3 array of pixels, although other output tile array sizes can also be used.
  • Such an output tile 54 may be stored in the look-up table 18 or, if the input tile 52 is processed using an algorithm as described in Fig. 2, the output tile 54 would be generated pursuant to a branch routine containing the exception algorithm.
  • Fig. 6 illustrates another input tile 56 exception which is a tick (pixel 57) abutting a row of pixels.
  • a tick is usually caused when the thickness of a line in an original document is very close to a threshold quantization level and only once in a while exceeds that threshold level. Thus, the tick is not a result of noise or a dirty original, but provides information as to the original image.
  • a tick is scaled in the corresponding output tile 58 to a thin horizontal line, equal to one-third the thickness of a full line. This assumes the output tile consists of 3 x 3 array of pixels. Thus, the thickness of the printed line containing the tick would increase slightly when the complete image has been printed.
  • Fig. 7 illustrates an input tile 60 containing a notch (pixel 61) in a line of pixels. This exception presumes that the notched line is slightly thinner than a full line and thus scales the notched line in the output tile 62 to a line equal to two-thirds the thickness of a full line. This assumes the output tile consists of a 3 x 3 array of pixels. The corrections to ticks and notches may also apply to vertical lines as well.
  • the vertical scaling of the printed image is 3.0 and the horizontal scaling is 1.5, so that three rows and three columns of pixels are outputted for every two horizontally adjacent original target pixels.
  • an original facsimile image is determined to be a picture (or other non-text image) in step 2 of Figs. 2 and 4, then it may not be desirable to correct this image since it is not known whether any features or angles are intentional or a result of quantization error.
  • it is determined that an input tile is a portion of an image no correction should be performed, and the target pixels in the input tile should be scaled to provide an output tile at the increased horizontal and vertical dpi of U,V, as shown in Fig. 2.
  • Fig. 8 is a flowchart of one method to determine whether the pixel data forms text or an image (e.g., a picture).
  • the determination that a received input tile in step 1 is part of an image is made if one or both of the following two conditions are met.
  • the first condition is if the number of exposed pixel faces (i.e., black-white transitions) within an input tile is greater than a certain threshold. If so, it is apparent that the minority pixels in the input tile are scattered and are part of an image (e.g., a halftone image). This is illustrated in steps 2, 3, and 4.
  • the second condition is if the longest black horizontal run is one pixel in all of the rows of the input tile, or the longest white horizontal run of pixels is one pixel in all of those rows (step 5). This second condition is consistent with the formation of very narrow lines or gaps which are used in pictures but not used in text. If the above conditions are met, the smoothing steps 5-7 in Fig. 2 are skipped, and the image data is processed in accordance with the remainder of the steps shown in Fig. 2. If a look-up table 18 for outputting enhanced output files is used, the image data may bypass the look-up table 18 and be scaled accordingly to match the dpi of the printer.
  • Figs. 2 or 4 for generating a modified output tile for that input tile would proceed (step 6 in Fig. 8).
  • the processor 14 in Fig. 1 would contain the necessary circuitry to determine whether the input tile pattern meets one or both of the above conditions.
  • suitable circuitry may be a comparator circuit which compares the input tile pattern to a number of stored patterns which meet the above two conditions.
  • One skilled in the art could easily fabricate such a circuit or develop an algorithm for making such a determination.
  • facsimile machines may be set to either a standard mode (98 x 203 dpi) or a fine mode (196 x 203 dpi).
  • a standard mode if it were desired to print at 300 dpi in both the vertical and horizontal directions, vertical scaling must be 3.0 and horizontal scaling must be 1.5.
  • the resulting three-row output tiles, containing nine pixels each, would require 8192 bytes in look-up table 18.
  • two-row output tiles containing six bits each would be used and require 4096 bytes in look-up table 18.
  • vertical scaling is 1.5, so only one or two-row output tiles are required, using only 4096 bytes in the look-up table 18 for each set of output tiles.
  • odd numbered rows of pixels may be scaled vertically by 1.0, and even rows of pixels may be scaled vertically by 2.0.
  • the four scaling tables mentioned above only consume 20 Kbytes to accommodate both the standard mode and fine mode in conventional facsimile machines.
  • circuitry of Fig. 1 and the algorithms described with respect to Figs. 2-4 may be implemented by one skilled in the art in view of this disclosure using only conventional circuitry and programming techniques.
  • the pixels generated using the above methods may be ultimately displayed on any medium, such as on paper or on a display screen.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)
  • Facsimile Image Signal Circuits (AREA)
EP95301537A 1994-03-11 1995-03-09 Pixelkorrektur und Glättungsverfahren Expired - Lifetime EP0672999B1 (de)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US08/212,438 US5537495A (en) 1994-03-11 1994-03-11 Pixel correction and smoothing method
US212438 1994-03-11

Publications (2)

Publication Number Publication Date
EP0672999A1 true EP0672999A1 (de) 1995-09-20
EP0672999B1 EP0672999B1 (de) 1999-08-04

Family

ID=22791013

Family Applications (1)

Application Number Title Priority Date Filing Date
EP95301537A Expired - Lifetime EP0672999B1 (de) 1994-03-11 1995-03-09 Pixelkorrektur und Glättungsverfahren

Country Status (4)

Country Link
US (1) US5537495A (de)
EP (1) EP0672999B1 (de)
JP (1) JP3765847B2 (de)
DE (1) DE69511136T2 (de)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6278804B1 (en) * 1996-11-29 2001-08-21 Kabushiki Kaisha Toshiba Image processing apparatus with smoothing feature
FR2871611A1 (fr) * 2004-06-15 2005-12-16 Siemens Vdo Automotive Sas Procede de mise a l echelle d'une image numerique en mode point

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3634419B2 (ja) * 1994-07-25 2005-03-30 セイコーエプソン株式会社 画像処理方法および画像処理装置
US5966475A (en) * 1997-02-20 1999-10-12 Hewlett-Packard Company System for enlarging and smoothing textual characters
US6563964B1 (en) * 1999-02-08 2003-05-13 Sharp Laboratories Of America, Inc. Image downsampling using redundant pixel removal
JP4134456B2 (ja) * 1999-09-17 2008-08-20 富士ゼロックス株式会社 多色画像形成装置
US6927780B2 (en) * 2002-01-14 2005-08-09 Seiko Epson Corporation Fast text/graphics resolution improvement with chain-code table look-up
JP4125245B2 (ja) * 2003-01-23 2008-07-30 キヤノン株式会社 文書表示方法および文書表示装置
TWI228240B (en) * 2003-11-25 2005-02-21 Benq Corp Image processing method for reducing jaggy-effect
US7652795B2 (en) * 2005-12-06 2010-01-26 Eastman Kodak Company Reducing ink bleed artifacts
JP2010193028A (ja) * 2009-02-17 2010-09-02 Seiko Epson Corp 画像処理のための装置、方法、プログラム
KR20130001213A (ko) * 2010-01-28 2013-01-03 이섬 리서치 디벨러프먼트 컴파니 오브 더 히브루 유니버시티 오브 예루살렘 엘티디. 입력 이미지로부터 증가된 픽셀 해상도의 출력 이미지를 생성하는 방법 및 시스템
CN109827585B (zh) * 2019-02-20 2022-08-02 哈尔滨工程大学 一种快速填补栅格地图中凹形区域的方法
CN111462147B (zh) * 2020-04-30 2022-07-05 柳州智视科技有限公司 一种基于图像块外轮廓及其角点对图像块进行切割和填补的方法

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5297244A (en) * 1992-09-22 1994-03-22 International Business Machines Corporation Method and system for double error antialiasing in a computer display system
EP0610928A2 (de) * 1993-02-12 1994-08-17 Fuji Xerox Co., Ltd. Prozessor zur Vergrösserung/Glättung von Bildern

Family Cites Families (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4193092A (en) * 1978-06-21 1980-03-11 Xerox Corporation Image interpolation system
US4280144A (en) * 1979-12-03 1981-07-21 International Business Machines Corporation Coarse scan/fine print algorithm
US4389677A (en) * 1980-12-08 1983-06-21 Ncr Canada Ltd - Ncr Canada Ltee Method and apparatus for removing erroneous elements from digital images
US4437122A (en) * 1981-09-12 1984-03-13 Xerox Corporation Low resolution raster images
US4578713A (en) * 1984-07-20 1986-03-25 The Mead Corporation Multiple mode binary image processing
US4742399A (en) * 1985-11-01 1988-05-03 Dainippon Screen Mfg. Co., Ltd. Method of and apparatus for generating high-resolution bilevel image data
US5001767A (en) * 1987-11-30 1991-03-19 Kabushiki Kaisha Toshiba Image processing device
JPH0832469B2 (ja) * 1988-01-29 1996-03-29 富士通株式会社 ビットマップフォントの横拡大方式
US5161213A (en) * 1988-05-27 1992-11-03 Wang Laboratories, Inc. Method for black and white image reduction based upon averaging black/white pixel counts of neighboring blocks
US5005139A (en) * 1988-08-16 1991-04-02 Hewlett-Packard Company Piece-wise print image enhancement for dot matrix printers
US4847641A (en) * 1988-08-16 1989-07-11 Hewlett-Packard Company Piece-wise print image enhancement for dot matrix printers
EP0356224B1 (de) * 1988-08-25 1996-12-11 Canon Kabushiki Kaisha Datenverarbeitungsgerät
JPH0286369A (ja) * 1988-09-22 1990-03-27 Toshiba Corp 画像処理装置
JP2813196B2 (ja) * 1989-04-12 1998-10-22 株式会社リコー 画像処理装置
US4975785A (en) * 1989-08-04 1990-12-04 International Business Machines Corporation Pel resolution addressing conversion
US4967203A (en) * 1989-09-29 1990-10-30 Hewlett-Packard Company Interlace printing process
US5025325A (en) * 1989-10-13 1991-06-18 Hewlett-Packard Company Graphics scaling method for high resolution printers
JP2647226B2 (ja) * 1990-03-23 1997-08-27 沖電気工業株式会社 ドットパターン圧縮方法及び伸長方法
JP2861251B2 (ja) * 1990-05-12 1999-02-24 ブラザー工業株式会社 印字装置
JP2502175B2 (ja) * 1990-08-18 1996-05-29 富士通株式会社 原画像パタ―ンデ―タ再生方法及び装置
US5029108A (en) * 1990-09-24 1991-07-02 Destiny Technology Corporation Edge enhancement method and apparatus for dot matrix devices
JP2510353B2 (ja) * 1990-11-07 1996-06-26 村田機械株式会社 画像デ―タのスム―ジング処理方法
US5404411A (en) * 1990-12-27 1995-04-04 Xerox Corporation Bitmap-image pattern matching apparatus for correcting bitmap errors in a printing system
JP2832089B2 (ja) * 1991-01-28 1998-12-02 キヤノン株式会社 画像処理装置
US5381522A (en) * 1991-03-28 1995-01-10 Canon Kabushiki Kaisha Image processing apparatus and method
US5270728A (en) * 1991-04-17 1993-12-14 Hewlett-Packard Company Raster imaging device speed-resolution product multiplying method and resulting pixel image data structure
US5267332A (en) * 1991-06-19 1993-11-30 Technibuild Inc. Image recognition system
JP3003318B2 (ja) * 1991-09-18 2000-01-24 セイコーエプソン株式会社 シリアルプリンタ
US5258854A (en) * 1991-12-06 1993-11-02 Xerox Corporation Converting between write-white, write-black and neutral bitmaps

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5297244A (en) * 1992-09-22 1994-03-22 International Business Machines Corporation Method and system for double error antialiasing in a computer display system
EP0610928A2 (de) * 1993-02-12 1994-08-17 Fuji Xerox Co., Ltd. Prozessor zur Vergrösserung/Glättung von Bildern

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
CORDINNIER: "antialiasing characters by pattern recognition", PROCEEDINGS OF THE SID, vol. 30, no. 1, USA, pages 23 - 28, XP000115922 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6278804B1 (en) * 1996-11-29 2001-08-21 Kabushiki Kaisha Toshiba Image processing apparatus with smoothing feature
FR2871611A1 (fr) * 2004-06-15 2005-12-16 Siemens Vdo Automotive Sas Procede de mise a l echelle d'une image numerique en mode point

Also Published As

Publication number Publication date
JP3765847B2 (ja) 2006-04-12
US5537495A (en) 1996-07-16
JPH07283946A (ja) 1995-10-27
EP0672999B1 (de) 1999-08-04
DE69511136D1 (de) 1999-09-09
DE69511136T2 (de) 1999-11-25

Similar Documents

Publication Publication Date Title
US5440407A (en) Pixel correction and smoothing method
US20020126313A1 (en) Image processing method and apparatus for varibly processing image based upon image characteristics
US5537495A (en) Pixel correction and smoothing method
US5483355A (en) Pixel correctional and smoothing method
US5768440A (en) Adaptive noise removal for video images with less correction of current pixel for higher variations between surrounding pixels
EP0810769B1 (de) Verfahren und Vorrichtung zum selektiven Bearbeiten von abgetasteten Bildvorlagen
US20030152283A1 (en) Image correction device, image correction method and computer program product in memory for image correction
JP2008011267A (ja) 画像処理装置、画像処理方法、画像処理プログラム及び記憶媒体
JP2000278523A (ja) 画像処理装置およびこれを搭載した画像読取装置と画像形成装置、並びに画像処理方法、および画像処理手順を記憶したコンピュータ読み取り可能な記憶媒体
JP3357776B2 (ja) 画像処理方法
JP2777202B2 (ja) 画像処理装置
EP0697783B1 (de) Punktdruck zwecks verbesserter Graphik
JP4050382B2 (ja) 多値画像のスムージング装置及び画像形成装置
JP3673675B2 (ja) 画像処理方法及び画像処理装置並びに記録媒体
JP3685148B2 (ja) 画像処理装置及び方法
JPH08307669A (ja) 画像処理装置及び方法
JPH01227576A (ja) 画像処理装置
JP2003230009A (ja) 画像形成装置
JP3472788B2 (ja) イメージ形成装置
JPH10276331A (ja) 画像処理方法および装置
JP3646694B2 (ja) 画像処理装置および方法
JPH10336454A (ja) 画像データ処理装置
JP4386870B2 (ja) 画像処理方法、画像処理装置、画像形成装置、コンピュータプログラム及び記録媒体
JP2002247351A (ja) 画像処理装置
JP2000156780A (ja) 画像処理装置

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): DE FR GB IT

17P Request for examination filed

Effective date: 19951228

17Q First examination report despatched

Effective date: 19980604

GRAG Despatch of communication of intention to grant

Free format text: ORIGINAL CODE: EPIDOS AGRA

GRAG Despatch of communication of intention to grant

Free format text: ORIGINAL CODE: EPIDOS AGRA

GRAH Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOS IGRA

GRAH Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOS IGRA

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): DE FR GB IT

REF Corresponds to:

Ref document number: 69511136

Country of ref document: DE

Date of ref document: 19990909

ITF It: translation for a ep patent filed
ET Fr: translation filed
PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

26N No opposition filed
REG Reference to a national code

Ref country code: GB

Ref legal event code: 732E

REG Reference to a national code

Ref country code: FR

Ref legal event code: TP

REG Reference to a national code

Ref country code: GB

Ref legal event code: IF02

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: IT

Payment date: 20060331

Year of fee payment: 12

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: GB

Payment date: 20080327

Year of fee payment: 14

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: FR

Payment date: 20080317

Year of fee payment: 14

Ref country code: DE

Payment date: 20080430

Year of fee payment: 14

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IT

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20070309

GBPC Gb: european patent ceased through non-payment of renewal fee

Effective date: 20090309

REG Reference to a national code

Ref country code: FR

Ref legal event code: ST

Effective date: 20091130

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20091001

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GB

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20090309

Ref country code: FR

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20091123